Category: Uncategorized
United Kingdom fact of the day
Britain recorded 65,000 more deaths than usual in the past three months as the coronavirus ravaged the country but numbers are now returning to normal, new data showed Tuesday.
In the week to June 19, so-called excess deaths in England and Wales fell below the five-year average for the first time since mid-March, the Office for National Statistics (ONS) said.
There were 0.7 percent fewer deaths overall than would be expected for that period.
Here is the full story.
Frequent, Fast, and Cheap is Better than Sensitive
A number of firms have developed cheap, paper-strip tests for coronavirus that report results at-home in about 15 minutes but they have yet to be approved for use by the FDA because the FDA appears to be demanding that all tests reach accuracy levels similar to the PCR test. This is another deadly FDA mistake.
NPR: Highly accurate at-home tests are probably many months away. But Mina argues they could be here sooner if the FDA would not demand that tests for the coronavirus meet really high accuracy standards of 80 percent or better.
A Massachusetts-based startup called E25Bio has developed this sort of rapid test. Founder and Chief Technology Officer Irene Bosch says her firm has field-tested it in hospitals. “What we learned is that the test is able to be very efficient for people who have a lot of virus,” she says.
The PCR tests can discover virus at significantly lower concentration levels than the cheap tests but that extra sensitivity doesn’t matter much in practice. Why not? First, at the lowest levels that the PCR test can detect, the person tested probably isn’t infectious. The cheap test is better at telling whether you are infectious than whether you are infected but the former is what we need to know to open schools and workplaces. Second, the virus grows so quickly that the time period in which the PCR tests outperforms the cheap test is as little as a day or two. Third, the PCR tests are taking days or even a week or more to report which means the results are significantly outdated and less actionable by the time they are reported.
The fundamental issue is this: if a test is cheap and fast we shouldn’t compare it head to head against the PCR test. Instead, we should compare test regimes. A strip test could cost $5 which means you can do one per day for the same price as a PCR test (say $35). Thus, the right comparison is seven cheap tests with one PCR test. So considered a stylized example. If a person gets infected on Sunday and is tested on Sunday then both tests will likely show negative. With the PCR test the infected person then goes to work, infecting other people throughout the week before being the person is tested again next Sunday. With the cheap test the person gets tested again on Monday and again comes up negative and they go to work but probably aren’t infectious. They are then tested again on Tuesday and this time there is enough virus in the person’s system to show positive so on Tuesday the infected person stops going to work and doesn’t infect anyone else. Score one for cheap tests. Now consider what happens if the person gets tested on another day, say Tuesday? In this case, both tests will show positive but the person doesn’t get the results of the PCR test until next Tuesday and so again goes to work and infects other people throughout the week. With the cheap test the infected person learns they are infected and again stops going to work and infecting other people. Score two for cheap tests.
Indeed, when you compare testing regimes it’s hard to come up with a scenario in which infrequent, slow, and expensive but very sensitive is better than frequent, fast, and cheap but less sensitive.
More details in this paper.
What should I ask Alex Ross?
I will be doing a Conversation with him. If you don’t know he writes for The New Yorker as a music (and literary) critic, writes a wonderful music blog, has first-rate books on music and has a new book coming out titled Wagnerism: Art and Politics in the Shadow of Music.
So what should I ask him?
Thursday assorted links
The case against the import of GPT-3
From an email from Agustin Lebron, noting that I will impose no further indentation:
“One thing that’s worth noting:
The degree of excitement about GPT-3 as a replacement for human workers, or as a path to AGI, is strongly inversely correlated with:
(a) How close the person is to the actual work. If you look at the tweets from Altman, Sutskever and Brockman, they’re pumping the brakes pretty hard on expectations.
(b) How much the person has actually built ML systems.
It’s a towering achievement to be able to train a system this big. But to me it’s clearly a dead-end on the way to AGI:
– The architecture itself is 3 years old: https://arxiv.org/abs/1706.03762. It is not an exaggeration to say that GPT-3’s architecture can be described as “take that 2017 paper and make 3 numbers (width, # layers, # heads) much bigger”. The fact that there hasn’t been any improvement in architecture in 3 years is quite telling.
– In the paper itself, the authors clearly say they’re quite near fundamental limits in being able to train an architecture like this. GPT-3 isn’t a starting point, it’s an end-point.
– If you look at more sober assessments (http://lacker.io/ai/2020/07/06/giving-gpt-3-a-turing-test.html, https://minimaxir.com/2020/07/gpt3-expectations/), without the tweet selection bias, it starts to look less impressive.
– Within my fairly heterogeneous circle of ML-expert friends, there’s little disagreement about dead-end-ness.
The most interesting thing about GPT-3 is the attention and press that it’s gotten. I’m still not sure what to make of that, but it’s very notable.
Again, it’s incredibly impressive and also piles of fun, but I’m willing to longbet some decent money that we’re not replacing front-end devs with attention-layer-stacks anytime soon.”
I remain bullish, but it is always worth considering other opinions.
Wednesday assorted links
1. Conversations on Christianity and liberalism.
2. Weird and clever sports plays, seven-minute video.
4. Russell Napier worries about inflation and financial repression.
5. B cell immunity.
6. David Autor, et.al., evaluate PPP.
7. Can insurance companies help reform police departments?
8. Andrew Gelman defends negativity.
The case for GPT-3
That is the topic of my latest Bloomberg column, here is one excerpt:
As a very rough description, think of GPT-3 as giving computers a facility with words that they have had with numbers for a long time, and with images since about 2012.
The core of GPT-3, which is a creation of OpenAI, an artificial intelligence company based in San Francisco, is a general language model designed to perform autofill. It is trained on uncategorized internet writings, and basically guesses what text ought to come next from any starting point. That may sound unglamorous, but a language model built for guessing with 175 billion parameters — 10 times more than previous competitors — is surprisingly powerful.
The eventual uses of GPT-3 are hard to predict, but it is easy to see the potential. GPT-3 can converse at a conceptual level, translate language, answer email, perform (some) programming tasks, help with medical diagnoses and, perhaps someday, serve as a therapist. It can write poetry, dialogue and stories with a surprising degree of sophistication, and it is generally good at common sense — a typical failing for many automated response systems. You can even ask it questions about God.
Imagine a Siri-like voice-activated assistant that actually did your intended bidding. It also has the potential to outperform Google for many search queries, which could give rise to a highly profitable company.
GPT-3 does not try to pass the Turing test by being indistinguishable from a human in its responses. Rather, it is built for generality and depth, even though that means it will serve up bad answers to many queries, at least in its current state. As a general philosophical principle, it accepts that being weird sometimes is a necessary part of being smart. In any case, like so many other technologies, GPT-3 has the potential to rapidly improve.
There is much more at the link.
*Counterpart* (no significant spoilers)
Girardian television! The basic set-up is that two similar, almost identical universes have branched from one, centered around Berlin. And there is a (controlled and limited) path for tunneling from one world to another. Some interaction ensues. Solve for the equilibrium. The Girardian equilibrium. It joins David Cronenberg’s Dead Ringers as one of the true cultural explanations of Girardian thought, and yes there is an embedded and I would say largely true model in the plot, especially in season two. Here is Wikipedia on the show. It was too smart to last for more than two years on the air. Excellent cast also, with the Whiplash drum instructor (J.K. Simmons) as the lead character(s), and the female lead from Rushmore (Olivia Williams) as his wife(s).
Tuesday assorted links
1. How to get people to wear masks. Don’t be rude to them or scold, and express your own vulnerability.
2. Further T-cell immunity results.
3. Proteus becomes the world’s first manufactured non-cuttable material.
4. Economics of Costco brands and their cannabalization.
5. Should we cancel Aristotle? (NYT)
6. Such a high seroprevalence in Delhi so quickly? And what is their actual death rate?
7. If you are hoping the second wave won’t be so bad, as we all are, Louisiana and Luxembourg are two bad news cases to worry about.
Moskos on Surging Crime in NYC
Peter Moskos warns about rising crime in NYC in the NYDaily news.
Violence in New York is up….In the last 28 days (through July 12), compared to last year, shootings have more than tripled (318 vs. 97). Last week was even worse. If the last 28 days become the new normal, 2021 will have more than 4,100 shootings, a level not seen in well over 20 years.
Undoubtedly bail reform, protests, looting, COVID-19, and the release of prisoners because of COVID all play a role, though how much is debate. What’s less known is how the NYPD has been methodically declawed by design.
Years of political advocacy have resulted in the intentional erosion of legal police authority. There is less prosecution. Most miscreant activities have been decriminalized. The city survived and even benefited from many reforms, but now the camel’s back is breaking.
…For many, this is a feature, not a flaw. A new breed of progressive prosecutors has battled to see who can prosecute the least. As a result, arrests in 2019 decreased 35% from 2016. Reducing incarceration is desirable, and New York has been doing so literally for decades without jeopardizing public safety.
More recently, since November, because of bail reform and COVID releases, the number of jailed inmates dropped another 40%. People are coming out of jail, and few are going in. Many applaud this because incarceration disproportionately affects Black and Brown people.
But so does non-enforcement and the rise in violence. In 2018 (the latest year with published data), 95.7% of shooting victims in New York City are Black or Hispanic. Just 4.3% of victims are white or Asian. When violence goes up, more Black and Hispanic people are shot.
How Swiss politics works
And now it’s becoming clear why almost all popular initiatives are rejected. If the initiative had a obvious chance of being approved, the parliament would introduce the necessary legislation on its own. From this point of view the small number of successful initiatives is not a sign of a system malfunction, but rather a proof that the system is functioning the way it is expected to.
And:
Another safety measure is that Swiss referenda are, in their essence, not polarizing. In referendum you are never asked to decide between two extremes, between, say, pro-life and pro-choice, but rather between the initiative proposal and the status quo. Voting against is always a safe and neutral option. It doesn’t necessarily mean that you are not sympathetic to the spirit of the initiative. You may just think it’s going too far, or maybe you like some aspects of it but don’t like some other.
Here is more from Martin Sustrik, via The Browser (always excellent). I’ll say it again: there should be far more books and articles asking the basic question of why Switzerland seems to work so well — Progress Studies!
The economy of Lebanon is collapsing
Here is the opening:
These are among the latest symptoms of an economic implosion that is accelerating at an alarming pace in Lebanon as its government, its banks and its citizens run out of foreign currency simultaneously.
And:
The Lebanese pound has lost over 60 percent of its value in just the past month, and 80 percent of its value since October. Prices are soaring and goods disappearing.
Bread, a staple of the Lebanese diet, is in short supply because the government can’t fund imports of wheat. Essential medicines are disappearing from pharmacies. Hospitals are laying off staff because the government isn’t paying its portion, and canceling surgeries because they don’t have electricity or the fuel to operate generators.
And:
Newly impoverished people are taking to Facebook to offer to trade household items for milk. Crime is on the rise. In one widely circulated video, a man wearing a coronavirus mask and wielding a pistol holds up a drugstore and demands that the pharmacist hand over diapers.
“Lebanon is no longer on the brink of collapse. The economy of Lebanon has collapsed,” said Fawaz Gerges, professor of international relations at the London School of Economics. “The Lebanese model established since the end of the civil war in 1990 has failed. It was a house of glass, and it has shattered beyond any hope of return.”
And:
Lebanon’s Western allies long ago made it clear that they won’t help out until the government undertakes efforts to reform the corrupt and bloated public sector. An $11 billion package of loans and investments has been on offer since 2018 — on the condition that the government undertake some limited changes. It hasn’t.
And if you had any doubts:
Staggering amounts are now missing from the banking system — perhaps as much as $100 billion, according to government figures.
Three-quarters of the deposits in the entire banking system were denominated in U.S. dollars, and many ordinary Lebanese may have lost most or all of their savings, said Jad Chaaban, an economist at the American University of Beirut.
Here is the full article by Liz Sly.
Monday assorted links
A highly speculative version of the immunological dark matter hypothesis
The COVID-19 pandemic is thought to began in Wuhan, China in December 2019. Mobility analysis identified East-Asia and Oceania countries to be highly-exposed to COVID-19 spread, consistent with the earliest spread occurring in these regions. However, here we show that while a strong positive correlation between case-numbers and exposure level could be seen early-on as expected, at later times the infection-level is found to be negatively correlated with exposure-level. Moreover, the infection level is positively correlated with the population size, which is puzzling since it has not reached the level necessary for population-size to affect infection-level through herd immunity. These issues are resolved if a low-virulence Corona-strain (LVS) began spreading earlier in China outside of Wuhan, and later globally, providing immunity from the later appearing high-virulence strain (HVS). Following its spread into Wuhan, cumulative mutations gave rise to the emergence of an HVS, known as SARS-CoV-2, starting the COVID-19 pandemic. We model the co-infection by an LVS and an HVS and show that it can explain the evolution of the COVID-19 pandemic and the non-trivial dependence on the exposure level to China and the population-size in each country. We find that the LVS began its spread a few months before the onset of the HVS and that its spread doubling-time is \sim1.59\pm0.17 times slower than the HVS. Although more slowly spreading, its earlier onset allowed the LVS to spread globally before the emergence of the HVS. In particular, in countries exposed earlier to the LVS and/or having smaller population-size, the LVS could achieve herd-immunity earlier, and quench the later-spread HVS at earlier stages. We find our two-parameter (the spread-rate and the initial onset time of the LVS) can accurately explain the current infection levels (R^2=0.74); p-value (p) of 5.2×10^-13). Furthermore, countries exposed early should have already achieved herd-immunity. We predict that in those countries cumulative infection levels could rise by no more than 2-3 times the current level through local-outbreaks, even in the absence of any containment measures. We suggest several tests and predictions to further verify the double-strain co-infection model and discuss the implications of identifying the LVS.
That is a new paper from Hagai and Ruth Perets, another link here, via Yaakov.
Sunday assorted links
1. Ross Douthat on meritocracy (NYT). And Caleb Watney on American innovation slowing (Atlantic).
2. Superspreading events. Very good piece.
4. What about single-strip testing yourself every day?
5. Parrondo’s paradox: how a combination of losing strategies (sometimes) can help you win. And an application to pandemics.
6. Too few low-wage jobs after Covid? And rolling 7-day averages for Covid deaths, worth checking that page regularly and don’t forget Sweden.